Abstract Many children with cerebral palsy (CP) show impairments in trunk postural control, which significantly impact their walking capacity and daily activities. For instance, children with severe CP, who have difficulties sitting independently, show poor directional specificity, with antagonists activating before agonists, which is distinct from typically developed children. Compared to typically developed peers, children with CP have a large range of motion for pelvis tilt, thorax, head, and kyphosis and lordosis during gait, even for some high functioning children with CP. While the significance of trunk motor control dysfunction in children with CP has been recognized, effective interventions for this core deficit are still lacking. Children with CP often receive or participate in a wide range of passive and active interventions aimed to improve postural control, but results have shown that current intervention approaches are not often effective in improving postural control in children with CP. For instance, hippotherapy, an intervention strategy that applies rhythmical force perturbations to the pelvis during sitting astride using horseback movement, has been used for improving balance and gait in children with CP for decades. However, while some studies showed improvements in balance and motor function in children with CP after hippotherapy, other studies indicated a mixed result regarding the effect of hippotherapy on Gross Motor Function Measure scores in children with CP. Thus, there is a critical need to improve the efficacy of current interventions for improving trunk postural control and gait in children with CP, which requires a thorough examination of the underlying neuromuscular mechanisms of the interventions. Our long-term goal is to develop rational-based intervention strategies to improve trunk postural control and gait in children with CP. The overall objective of this study is to examine the neuromuscular mechanisms of the trunk muscles to a force perturbation applied to the pelvis during sitting astride, and determine whether repeated exposure to pelvic perturbations during sitting astride using a robotic system will be effective in improving trunk postural control and gait in children with CP. Our central hypothesis is that repetitive activation of specific sensorimotor pathways through applying targeted force perturbations will improve postural control in children with CP, which may be due to the reinforcement in circuits and synapses used for trunk postural control through a use-dependent neuroplasticity mechanism. The rationale for the proposed study is that an understanding of the neuromuscular mechanisms of trunk muscle control in children with CP and determination of the therapeutic effect of targeted force perturbation are likely to provide a strong scientific foundation whereby new force perturbation based intervention strategies can be developed to improve postural control and gait in children with CP. The results from this study may be used to develop innovative clinical therapies aimed at improving trunk postural control and walking function in children with CP.